Statistics and Mishaps

The theme lately has been small mishaps that in the solitude of "bachelorhood" seem like a big deal (at least for a little while). First one was me trying out my "Food Network" skills when I needed to flip over an egg I was frying. My one spatula was dirty so I thought instead I would just flip it using the pan the way I see all the cooks on TV. I have done it before with other food items so it seemed like the right move. Well, I missed judged the trajectory and the egg ended up flush on the burner. At least I was smart enough to NOT reach to try to catch it or grab it off the burner. Cremated eggs linger a lot longer than I would have figured. It has been cleaned up now, but there seems to be some iota that remains and relights each time I use that burner (it is only large one I have. The second little mishap came on my first workout in the apartment complex pool. A little back story. The brochure said the complex has an "Olympic Size Pool." Hotel and Campground signs that advertised such aquatic facilities were ubiquitous back in days of our family vacationing via travel trailer and not once did the pool measure up to the claim. Well in turns out that Easy Bay/Pelican Bay actually has a four lane, 25 yard pool. Not a Beijing Olympic Pool, but that is pretty close to the dimensions of the pool used for the 1904 games in St. Louis. In this election year, that counts as a lead pipe fact compared to some of the spin we have witnessed. Anyway, I manage to get in a very slow and lumbering 2000 yards before I got cocky and decided to do some backstroke. It has been a while since I have been in an outdoor pool with few sight lines and no backstroke flags so despite what I thought were my best efforts, I cracked my head on the wall on the second length of back. Fortunately I am no worse for the wear. The statistics goes with the dissertations I mention in the last post. Statistics, along with Calculus and being fluent in a foreign language are the big regrets of my life at this point. Anyway, I have spent the better part of the last two days trying to figure out a compromise for a student's statistical analysis that doesn't match the data he collected so that he can still answer the research questions he intended. I have found myself shoulders deep in the literature that features the dispute about whether replies to surveys with Likert Scale answers. These are the surveys that ask you to reply with things like Strongly Agree, Agree, Neutral, Disagree, Strongly Disagree, but they have you circle a number such as 1 - 5 instead. Long story short some statisticians argue that you can use those "numbers" like real integers and other argue that they are really categories and are not numbers that can be manipulated mathematically. I fall squarely with the latter group. The big problem is that many social scientists and most certainly graduate students have cast their lots with the former group; not because they understand what they are doing, but because the "saw someone else do it some where." As with most graduate student stories we are under a huge time crunch with really high stakes for completion. The solution clearly includes a fancy mathematical manipulation, the proofs of which are making my head hurt. I do think I have figured out a way to get all the pieces in place so we can finish him up. Tomorrow will tell. I am off to work on my spanish.